I have been dabbling in smart home technology and voice assistants for several years now. I was one of the first people I know to own an Amazon Echo (the original black tube), and used it until November 2016, when the Google Home first came out. Since I am an Android user, and I use Google’s services heavily in multiple areas of my life, I made the switch to the Google Home and never looked back. That is, until mid-2017, when the Mycroft Mark 1 Kickstarter was first announced.
Mycroft is an open source voice assistant, and the Mark 1 was the first hardware kit designed to run the Mycroft software. It consists of a Raspberry Pi 3, a speaker, a microphone, and a small LED display, all in a friendly white case. I chose not to back the project when it first came out since I was somewhat skeptical of the developers’ ability to deliver on their goals, but I continued to follow the project with interest.
Fast forward to early 2018, and the Mycroft team once again ran a Kickstarter campaign to finance a new hardware project. This time, the device was the Mycroft Mark II, and the system had matured to the point where I was comfortable backing it. I opted to back at the level where I would get both a Mark II when it comes out (est. December 2018), and a Mark 1 earlier in the year (est. April 2018).
All of that leads me to this week, when my Mycroft Mark 1 device finally arrived. Unfortunately, on the day that it showed up on my doorstep, I had to leave for a quick business trip, and so I didn’t have time to actually open anything until Saturday morning (today).
My first impressions are that, as expected, the Mark 1 is at best a reference device, good for developers and early adopters. That is, after all, what the Mark 1 was marketed as. It is nowhere near as polished as the Google Home, nor is it as useful to me currently. However, I don’t plan to let that stop me from continuing to use it. My current plans are to set it up in my home office, and tinker with various skills and integrations as they become available. One interesting integration already available is with Home Assistant, which is what I am currently using as the hub for all of my smart home gadgets. Also, if I have time over the summer, I may experiment with writing my own skills for Mycroft.
So, why spend time and money on a device that I knew from the start wouldn’t be as good as Google or Amazon’s offerings? Simple. While I am heavily invested in both the Google and Amazon ecosystems, I don’t actually like the idea of either one of those companies having microphones in every room of my house. I am continuing to use their products for now, since I am curious enough about new advances in the smart home and voice assistant spaces that I am willing to sacrifice some privacy. But, I want to foster competition in the marketplace, and the best option from a privacy standpoint is an opensource, self-hosted solution. Hopefully, given time and support, Mycroft will become that solution.